GSM8K-Binary_Llama-3.2-1B-cmatu09u

This model is a fine-tuned version of meta-llama/Llama-3.2-1B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3294
• Model Preparation Time: 0.0057
• Mdl: 4746.8979
• Accumulated Loss: 3290.2989
• Correct Preds: 1929.0
• Total Preds: 2475.0
• Accuracy: 0.7794
• Correct Gen Preds: 1897.0
• Gen Accuracy: 0.7665
• Correct Gen Preds 34192: 1020.0
• Correct Preds 34192: 1040.0
• Total Labels 34192: 1196.0
• Accuracy 34192: 0.8696
• Gen Accuracy 34192: 0.8528
• Correct Gen Preds 41568: 869.0
• Correct Preds 41568: 889.0
• Total Labels 41568: 1267.0
• Accuracy 41568: 0.7017
• Gen Accuracy 41568: 0.6859

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Mdl	Accumulated Loss	Correct Preds	Total Preds	Accuracy	Correct Gen Preds	Gen Accuracy	Correct Gen Preds 34192	Correct Preds 34192	Total Labels 34192	Accuracy 34192	Gen Accuracy 34192	Correct Gen Preds 41568	Correct Preds 41568	Total Labels 41568	Accuracy 41568	Gen Accuracy 41568
No log	0	0	1.4656	0.0057	5233.1723	3627.3586	1196.0	2475.0	0.4832	1204.0	0.4865	1196.0	1196.0	1196.0	1.0	1.0	0.0	0.0	1267.0	0.0	0.0
0.7705	1.0	20	0.7229	0.0057	2581.2769	1789.2048	1387.0	2475.0	0.5604	1395.0	0.5636	151.0	151.0	1196.0	0.1263	0.1263	1236.0	1236.0	1267.0	0.9755	0.9755
0.7361	2.0	40	0.8157	0.0057	2912.7554	2018.9682	1357.0	2475.0	0.5483	12.0	0.0048	0.0	105.0	1196.0	0.0878	0.0	3.0	1252.0	1267.0	0.9882	0.0024
0.7985	3.0	60	0.6055	0.0057	2161.9344	1498.5388	1805.0	2475.0	0.7293	9.0	0.0036	0.0	697.0	1196.0	0.5828	0.0	0.0	1108.0	1267.0	0.8745	0.0
1.027	4.0	80	0.6487	0.0057	2316.2232	1605.4836	1907.0	2475.0	0.7705	827.0	0.3341	504.0	1037.0	1196.0	0.8671	0.4214	315.0	870.0	1267.0	0.6867	0.2486
1.0135	5.0	100	0.8148	0.0057	2909.2559	2016.5426	1876.0	2475.0	0.7580	444.0	0.1794	267.0	1061.0	1196.0	0.8871	0.2232	169.0	815.0	1267.0	0.6433	0.1334
0.6827	6.0	120	1.0484	0.0057	3743.6020	2594.8672	1910.0	2475.0	0.7717	1146.0	0.4630	604.0	1013.0	1196.0	0.8470	0.5050	534.0	897.0	1267.0	0.7080	0.4215
0.494	7.0	140	1.1964	0.0057	4272.1202	2961.2080	1753.0	2475.0	0.7083	1133.0	0.4578	802.0	1130.0	1196.0	0.9448	0.6706	322.0	623.0	1267.0	0.4917	0.2541
0.0061	8.0	160	1.0888	0.0057	3887.6479	2694.7122	1891.0	2475.0	0.7640	1312.0	0.5301	750.0	1062.0	1196.0	0.8880	0.6271	554.0	829.0	1267.0	0.6543	0.4373
0.0001	9.0	180	1.2353	0.0057	4410.9305	3057.4240	1894.0	2475.0	0.7653	1628.0	0.6578	928.0	1059.0	1196.0	0.8855	0.7759	690.0	835.0	1267.0	0.6590	0.5446
0.0352	10.0	200	1.3294	0.0057	4746.8979	3290.2989	1929.0	2475.0	0.7794	1897.0	0.7665	1020.0	1040.0	1196.0	0.8696	0.8528	869.0	889.0	1267.0	0.7017	0.6859
0.0002	11.0	220	1.6740	0.0057	5977.3293	4143.1690	1873.0	2475.0	0.7568	1862.0	0.7523	1066.0	1070.0	1196.0	0.8946	0.8913	788.0	803.0	1267.0	0.6338	0.6219
0.0	12.0	240	1.6150	0.0057	5766.6040	3997.1053	1899.0	2475.0	0.7673	1890.0	0.7636	1039.0	1044.0	1196.0	0.8729	0.8687	843.0	855.0	1267.0	0.6748	0.6654
0.0	13.0	260	1.5999	0.0057	5712.7643	3959.7865	1901.0	2475.0	0.7681	1893.0	0.7648	1034.0	1039.0	1196.0	0.8687	0.8645	851.0	862.0	1267.0	0.6803	0.6717
0.0	14.0	280	1.5964	0.0057	5700.2985	3951.1459	1900.0	2475.0	0.7677	1893.0	0.7648	1032.0	1037.0	1196.0	0.8671	0.8629	853.0	863.0	1267.0	0.6811	0.6732
0.0	15.0	300	1.5956	0.0057	5697.2293	3949.0184	1900.0	2475.0	0.7677	1893.0	0.7648	1031.0	1036.0	1196.0	0.8662	0.8620	854.0	864.0	1267.0	0.6819	0.6740
0.0	16.0	320	1.5949	0.0057	5694.9231	3947.4199	1903.0	2475.0	0.7689	1895.0	0.7657	1031.0	1036.0	1196.0	0.8662	0.8620	856.0	867.0	1267.0	0.6843	0.6756
0.0	17.0	340	1.5931	0.0057	5688.4595	3942.9397	1903.0	2475.0	0.7689	1895.0	0.7657	1030.0	1035.0	1196.0	0.8654	0.8612	857.0	868.0	1267.0	0.6851	0.6764
0.0	18.0	360	1.5933	0.0057	5689.2287	3943.4728	1898.0	2475.0	0.7669	1892.0	0.7644	1028.0	1033.0	1196.0	0.8637	0.8595	856.0	865.0	1267.0	0.6827	0.6756
0.0	19.0	380	1.5946	0.0057	5693.9438	3946.7411	1898.0	2475.0	0.7669	1892.0	0.7644	1028.0	1033.0	1196.0	0.8637	0.8595	856.0	865.0	1267.0	0.6827	0.6756
0.0	20.0	400	1.5934	0.0057	5689.5182	3943.6735	1899.0	2475.0	0.7673	1893.0	0.7648	1026.0	1031.0	1196.0	0.8620	0.8579	859.0	868.0	1267.0	0.6851	0.6780
0.0	21.0	420	1.5928	0.0057	5687.2881	3942.1277	1900.0	2475.0	0.7677	1893.0	0.7648	1029.0	1033.0	1196.0	0.8637	0.8604	856.0	867.0	1267.0	0.6843	0.6756
0.0	22.0	440	1.5908	0.0057	5680.3600	3937.3255	1904.0	2475.0	0.7693	1897.0	0.7665	1028.0	1033.0	1196.0	0.8637	0.8595	861.0	871.0	1267.0	0.6875	0.6796
0.0	23.0	460	1.5912	0.0057	5681.5422	3938.1450	1903.0	2475.0	0.7689	1896.0	0.7661	1027.0	1032.0	1196.0	0.8629	0.8587	861.0	871.0	1267.0	0.6875	0.6796
0.0	24.0	480	1.5899	0.0057	5676.8926	3934.9221	1902.0	2475.0	0.7685	1895.0	0.7657	1026.0	1031.0	1196.0	0.8620	0.8579	861.0	871.0	1267.0	0.6875	0.6796
0.0	25.0	500	1.5925	0.0057	5686.1677	3941.3511	1898.0	2475.0	0.7669	1891.0	0.7640	1023.0	1028.0	1196.0	0.8595	0.8554	860.0	870.0	1267.0	0.6867	0.6788
0.0	26.0	520	1.5916	0.0057	5682.9516	3939.1219	1899.0	2475.0	0.7673	1893.0	0.7648	1023.0	1028.0	1196.0	0.8595	0.8554	862.0	871.0	1267.0	0.6875	0.6803
0.0	27.0	540	1.5918	0.0057	5683.6733	3939.6221	1906.0	2475.0	0.7701	1900.0	0.7677	1026.0	1031.0	1196.0	0.8620	0.8579	866.0	875.0	1267.0	0.6906	0.6835
0.0	28.0	560	1.5920	0.0057	5684.3380	3940.0829	1902.0	2475.0	0.7685	1896.0	0.7661	1024.0	1028.0	1196.0	0.8595	0.8562	864.0	874.0	1267.0	0.6898	0.6819
0.0	29.0	580	1.5914	0.0057	5682.5389	3938.8358	1905.0	2475.0	0.7697	1899.0	0.7673	1024.0	1029.0	1196.0	0.8604	0.8562	867.0	876.0	1267.0	0.6914	0.6843
0.6534	30.0	600	1.5917	0.0057	5683.5174	3939.5141	1903.0	2475.0	0.7689	1898.0	0.7669	1024.0	1028.0	1196.0	0.8595	0.8562	866.0	875.0	1267.0	0.6906	0.6835
0.6535	31.0	620	1.5925	0.0057	5686.4309	3941.5335	1904.0	2475.0	0.7693	1899.0	0.7673	1022.0	1027.0	1196.0	0.8587	0.8545	869.0	877.0	1267.0	0.6922	0.6859
0.0	32.0	640	1.5930	0.0057	5688.0845	3942.6797	1904.0	2475.0	0.7693	1898.0	0.7669	1024.0	1029.0	1196.0	0.8604	0.8562	866.0	875.0	1267.0	0.6906	0.6835
0.0	33.0	660	1.5921	0.0057	5684.7574	3940.3736	1908.0	2475.0	0.7709	1902.0	0.7685	1024.0	1028.0	1196.0	0.8595	0.8562	870.0	880.0	1267.0	0.6946	0.6867
0.0	34.0	680	1.5929	0.0057	5687.8234	3942.4988	1904.0	2475.0	0.7693	1898.0	0.7669	1023.0	1028.0	1196.0	0.8595	0.8554	867.0	876.0	1267.0	0.6914	0.6843
0.0	35.0	700	1.5931	0.0057	5688.4900	3942.9608	1907.0	2475.0	0.7705	1902.0	0.7685	1024.0	1029.0	1196.0	0.8604	0.8562	870.0	878.0	1267.0	0.6930	0.6867
0.0	36.0	720	1.5928	0.0057	5687.3227	3942.1517	1902.0	2475.0	0.7685	1896.0	0.7661	1021.0	1026.0	1196.0	0.8579	0.8537	867.0	876.0	1267.0	0.6914	0.6843

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1